252 B. GRASSI AND A. SANDIAS. 
body ; the latter is accompanied by the presence of a bacterium 
which I have not investigated. 
If the wood is too damp the colony may generally be saved 
by the simple precaution of leaving the cork out of the tube ; 
after a few days the moisture diminishes, and the cork must 
then be re-inserted, or the opposite extreme of undue dryness 
is quickly reached. With time and patience an exact estimate 
can be formed of the amount of moisture necessary, and it 
can then be easily regulated. The cork must fit closely, and 
want of oxygen need not be feared. At one time I was 
accustomed to uncork the tubes three or four times a day, 
but I subsequently found this to be quite unnecessary; if the 
Calotermites require change of air, they are able to provide it 
for themselves by boring through the cork. 
When these tiny nests are examined the tube may conve- 
niently be laid flat and left quiet, but an occasional shake is 
sometimes useful] to rouse the inmates to activity. 
Unfortunately Termes lucifugus does not flourish in 
these tubes, but drags on a feeble existence for a week or ten 
days at most; I have made many but quite unsuccessful 
attempts to establish a moderately suitable environment for 
them therein. The species is all the more difficult to study 
because, in comparison with Calotermes, the members of a 
colony are smaller and more rapid in movement. 
Many as have been the details made out through these glass 
tubes in my laboratory, there are others which will be more 
successfully undertaken by anyone who has the opportunity of 
applying my method to the large exotic species. 
Considering that the termites of tropical countries are 
among the most injurious of insects, I must point out a 
practical result of my investigations, and one of the highest 
importance. Contrary to the belief of residents and the 
accounts of travellers, such as Major Casati,! a nest of termites 
cannot be destroyed merely by killing the king and queen. 
If there remain alive some eight, ten, or twenty examples, 
which include any undifferentiated larve, or larve of perfect 
1 («Ten Years in Equatorial Africa,’ i, p. 165.] 
